Predictors for progression of two different types of cervical lesions in rheumatoid arthritis treated with biologic agents

Abstract

Background

Biologic agents (BAs) enabled not only a reduction of disease activity but also a slowing down of structural damage to the joints in patients with rheumatoid arthritis (RA). However, the incidence of cervical lesions in patients with RA is still high.

Purpose

To elucidate the predictors for the progression of two different cervical lesions in patients with RA under BA treatment.

Methods

Of 151 subjects who received more than two years of continuous BA treatment, 101 subjects who had cervical X-ray images taken at baseline and final visit were enrolled. The mean disease duration and mean radiography interval were 10.6 years and 4.4 years, respectively. The existence and progression of cervical lesions (atlanto-axial subluxation [AAS], vertical subluxation [VS], and subaxial subluxation [SS]) were investigated. And predictors for the AAS or VS progression were analyzed by multivariate logistic regression analysis.

Results

The incidence of cervical lesions at baseline were no pre-existing cervical lesion (none) in 50 cases (50%), AAS only in 32 (32%), both AAS and VS in 12 (12%), and VS only in 7 cases (7%). In the none group, only 4 cases of AAS progression (8%) was observed during the follow-up. In contrast, in the groups with pre-existing cervical lesions, a high incidence of VS progression was observed (63% in the AAS only group, 58% in the AAS + VS group, and 71% in the VS only group). Multivariate regression analysis demonstrated that the DAS-CRP value at baseline (odds ratio [OR] = 9.23) and matrix metaloprotease-3 level at baseline (OR = 1.01) were significant predictors for the progression of AAS, and pre-existing AAS (OR = 18.38) was a sole significant predictor for the progression of VS.

Conclusions

Cervical lesions progressed irrespective of disease activity after AAS development. Strict disease control before the development of AAS is crucial for preventing further progression and development of cervical lesions.

Introduction

Rheumatoid arthritis (RA) is the most prevalent cause of inflammatory arthritis, affecting 1% of the population [1]. Swelling and joint destruction at peripheral joints are frequent and conspicuous on appearance, but inconspicuous cervical spine involvement in patients with RA is also common (25%–86%) [2], [3], [4], [5], [6], [7]. Anatomic deformities caused by cervical lesions can cause spinal cord or brainstem compression, with resultant neurological deficits, such as cervical myelopathy, paresis, and even death [1]. Rheumatoid cervical lesions are classified into three types. Instability between C1-2, named atlantoaxial subluxation (AAS), is the most common type of cervical lesion, which is caused by laxity or failure of the transverse, apical, and alar ligaments due to rheumatoid inflammation. The second common type of cervical lesion is vertical subluxation (VS), which is characterized by cranial migration of the C2 odontoid process by joint destruction between C1-2 and occiput-C1. The least common type of lesion is subaxial subluxation (SS), which is anterior-posterior spinal instability caused by destruction of the facet joints as well as the intervertebral disc [8]. Treatment strategies for RA have recently undergone a major shift. Standard of care now entails initiating immediate treatment using methotrexate (MTX) or a combination of MTX plus biological agents (BAs), and the incidence of clinically relevant joint destruction has decreased [9], [10], [11]. The unique ligamentous stabilizing mechanism at C1-2, whose failure of function leads to AAS, may be more susceptible to the effect of inflammation and synovitis than bony joints at the C1-2 lateral mass or facet joints, which leads to VS and then SS. Recent studies have reported that the incidence of cervical lesions is still high (32%), even in patients with RA onset after 2000 [12] BAs can stave off the emergence of de novo cervical spine lesions, but they cannot prevent the progression of pre-existing cervical spine lesions [13]. However, in these studies, the predictors for the progression of respective cervical lesions were not analyzed independently. The purpose of this study was to elucidate the predictors for the progression of respective types of cervical lesions.